ACCT 210 Lecture Notes - Lecture 10: Cellulosome, Ramat Aviv, Elsevier
Cohesin–dockerin interaction in cellulosome assembly: a single
Asp-to-Asn mutation disrupts high-affinity cohesin–dockerin binding
Tal Handelsman
a
, Yoav Barak
b
, David Nakar
b
, Adva Mechaly
b
, Raphael Lamed
c
,
Yuval Shoham
a,d
, Edward A. Bayer
b,*
a
Department of Biotechnology and Food Engineering, Technion – Israel Institute of Technology, Haifa, Israel
b
Department of Biological Chemistry, The Weizmann Institute of Science, Rehovot, Israel
c
Department of Molecular Microbiology and Biotechnology, Tel-Aviv University, Ramat Aviv, Israel
d
Institute of Catalysis Science and Technology, Technion – Israel Institute of Technology, Haifa, Israel
Received 25 May 2004; revised 15 July 2004; accepted 16 July 2004
Available online 26 July 2004
Edited by Ulf-Ingo Fl€
ugge
Abstract The cohesive cellulosome complex is sustained by the
high-affinity cohesin–dockerin interaction. In previous work
[J. Biol. Chem. 276 (2001) 9883], we demonstrated that a single
Thr-to-Leu replacement in the Clostridium thermocellum dock-
erin component differentiates between non-recognition and high-
affinity recognition by the interspecies rival cohesin from
C. cellulolyticum. In this report, we show that a single Asp-to-
Asn substitution on the cohesin counterpart also disrupts normal
recognition of the dockerin. The Asp34 carboxyl group of the
cohesin appears to play a central role in the resultant hydrogen-
bonding network as an acceptor of two crucial hydrogen bonds
from Ser45 of the dockerin domain. The results underscore the
fragile nature of the intermolecular contact interactions that
maintain this very high-affinity protein–protein interaction.
Ó2004 Published by Elsevier B.V. on behalf of the Federation of
European Biochemical Societies.
Keywords: Cellulosome; Protein–protein interaction;
Cohesin–dockerin binding specificity; Multi-enzyme complex;
Clostridium thermocellum;Clostridium cellulolyticum
1. Introduction
Cellulosomes are multi-enzyme extracellular complexes,
produced by various anaerobic microorganisms for the effi-
cient degradation of plant cell wall polysaccharides [1–4]. The
various cellulosome components are assembled by virtue of a
high-affinity protein–protein interaction between reciprocal
modules on the interacting subunits – the cohesin and the
dockerin. In early studies on the cellulosomes from two clos-
tridial species, Clostridium thermocellum and C. cellulolyticum,
the interaction between cohesins and dockerins was found to
be generally species specific: experiments carried out with
isolated modules from the two species revealed that cohesins
from the scaffoldin of one species bind to the dockerins of its
own enzymatic subunits with high affinity, but fail to recognize
those of the other species despite the relatively high sequence
homology among the analogous components [5,6].
Crystal structures of cohesins from the scaffoldin of
C. thermocellum [7,8] and C. cellulolyticum [9] have been
published. The cohesins form a nine-stranded b-sandwich with
a jelly-roll topology. The b-sandwich results from the associ-
ation of a four-stranded antiparallel b-sheet and a five-stran-
ded mixed b-sheet, stabilized by a hydrophobic core. The two
b-sheets are composed of strands 8, 3, 6, 5 and strands 9, 1, 2,
7, 4, respectively. In addition, a solution structure of a dock-
erin from C. thermocellum cellulosomal cellobiohydrolase CelS
has been solved by NMR analysis [10]. The structure consists
of two Ca2þ-binding loop-helix motifs that bear sequence
homology to the EF-hand motif of eukaryotic calcium-binding
proteins, such as calmodulin and troponin C. Very recently, a
crystal structure of a cohesin–dockerin complex from C.
thermocellum has also been solved [11]. The complex shows
that, while the cohesin module remains essentially unchanged,
the dockerin undergoes conformational adjustments upon
binding. The protein–protein contact between one face of the
cohesin and a-helices 1 and 3 of the dockerin is mediated
mainly by hydrophobic interactions and relatively few inter-
molecular hydrogen bonds. Although the structure of the
heterodimer sheds additional light on the structural basis of
the cohesin–dockerin interface, the function and importance of
specific amino acids involved in recognition and binding are
not entirely apparent from the structural data.
To determine the exact role of the contact residues in
binding and affinity, various residues on the C. thermocellum
cohesin surface were replaced with matching residues of C.
cellulolyticum. The binding specificity and affinity of the re-
sultant mutated proteins were tested using an enzyme-linked
assay. Although the mutated cohesins maintained their origi-
nal specificity, a dramatic reduction in the affinity was ob-
served for several of the mutants, the common denominator
being the D34N mutation. This single conservative amino acid
replacement reduces the affinity of the interaction by more
than 3 orders of magnitude.
2. Materials and methods
2.1. Protein constructs and cloning
The protein construct containing the cohesin from C. thermocellum
consists of cohesin-2 and a cellulose-binding domain from CipA. The
construct containing the cohesin from C. cellulolyticum comprises a
cellulose-binding domain, a hydrophilic domain, and cohesin-1 from
CipC. Details of the cloning of cohesin constructs (termed Coh2CBD-t
*
Corresponding author. Fax: +972-8-9468256.
E-mail address: [email protected] (E.A. Bayer).
0014-5793/$22.00 Ó2004 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies.
doi:10.1016/j.febslet.2004.07.040
FEBS 28684 FEBS Letters 572 (2004) 195–200
and miniCipC-c, where t and c denote domains derived from C.
thermocellum and C. cellulolyticum, respectively) were described else-
where [5,12].
The dockerin constructs comprise the dockerin domain of CelS from
C. thermocellum [13] or the dockerin domain of CelA from C. cellul-
olyticum [14], fused downstream of the non-cellulosomal family-10
xylanase T-6 from Geobacillus stearothermophilus [15,16]. These con-
structs, termed XynDocS-t and XynDocA-c, respectively, were cloned
using a specially designed cassette produced for this purpose. The
cassette consisted of the gene for the G. stearothermophilus xylanase T-
6 with a His-tag and a BspHI site at the 50-terminus and a KpnI site at
the 30-terminus. This construct was ligated at the KpnI site with the
PCR product of a C. thermocellum CelS (Cel48A) dockerin (containing
a5
0-terminal KpnI site and a 30-terminal BamHI site) and inserted into
the pET9d vector at the NcoI and BamHI sites. This plasmid allows
replacement of the CelS dockerin with any other desired dockerin by
digesting with KpnI and BamHI, and the resultant expressed product
constitutes a His-tagged xylanase T-6 fusion-protein bearing a dock-
erin at the C-terminus.
2.2. Site-directed mutagenesis
The mutagenesis of the cohesin domain was carried out as previously
described [17]. Generally, mutated cohesins containing combined
mutations were produced in a sequential manner, in which one mutant
served as a template for the subsequent one.
2.3. Expression and purification of proteins
All proteins were expressed in Escherichia coli BL21(DE3) grown
overnight in Terrific Broth medium [18]. For the production of mini-
CipC-c cloned in pET22b, the medium was supplemented with 0.1 mg/
ml ampicillin, and protein expression was induced with 0.4 mM iso-
propyl b-
DD
-thiogalactoside. For all other proteins cloned in pET9d, the
medium was supplemented with 25 lg/ml kanamycin, and growth was
carried out without induction. Following growth, cells were harvested,
resuspended in TrisNC buffer (50 mM Tris, 100 mM NaCl, 2 mM
CaCl2, and 0.02% sodium azide, pH 7.5), disrupted by two passages
through a FrenchÒpress (Spectronic Instruments, Inc., Rochester,
NY, USA), and centrifuged for the production of clear crude protein
extracts, that were further purified as described below.
Xylanase-containing constructs (XynDocS-t and XynDocA-c) were
purified by gel filtration using a Superdex 200 26/60 column, AKTA
explorer (Pharmacia), running at 2.5 ml/min with TrisNC buffer. CBD-
containing constructs (Coh2CBD-t and Coh1-c) were purified by af-
finity chromatography on cellulose. Microcrystalline cellulose (Avicel
Type PH-101 FMC) was added to the crude protein extract, origi-
nating from a 1-l cell culture. The ratio of cellulose to cells was 0.7 g
per 1 unit OD600. The resultant suspension was stirred for 1 h. After
centrifugation, the pellet was washed twice with TrisNC buffer, con-
taining 0.1 M NaCl and twice with TrisNC buffer, containing 1 M
NaCl. The CBD-containing proteins were eluted from the cellulosic
matrix with 11 ml of 1% (v/v) triethylamine. The eluent fractions were
neutralized with TrisNC buffer. Purity of all proteins was estimated by
SDS–PAGE and protein concentration was estimated by Bradford
[19].
2.4. Non-competitive enzyme-linked interaction assay
Microtiter plates (MaxiSorp-immunoplates, NUNC A/S, Roskilde,
Denmark) were coated overnight at 23 °C with the cohesin test samples
(200 ll/well, 270 nM of miniCipC-c, wild-type or mutated Coh2CBD-
t). The plates were blocked for 2.5 h with blocking solution (300 ll/
well, 3% (w/v) of bovine serum albumin in TrisNC buffer) and washed
three times with TrisNC buffer (300 ll/well). The cohesin–dockerin
interaction was initiated upon addition of dockerin samples (200 ll/
well, 94 nM of XynDocA-c or XynDocS-t), and the plates were in-
cubated for 2.5 h. After five washes, the bound dockerins were detected
by means of the fused-xylanase activity: substrate solution (240 ll/well,
2.9 mM of p-nitrophenyl b-
DD
-cellobioside) was added followed by
incubation at 60 °C. Optical density was determined at 420 nm on a
VERSAmax microplate reader (Molecular Devices Corp., Sunnyvale
CA).
2.5. Competitive enzyme-linked interaction assay
Microtiter plates were coated overnight with wild-type C. thermo-
cellum cohesin samples (200 ll/well, 270 nM of Coh2CBD-t). Plates
were blocked for 2.5 h with the above-described blocking solution and
washed three times with TrisNC buffer. The cohesin–dockerin inter-
action was carried out by the addition of 100 ll of the desired com-
petitor cohesin sample (i.e., wild-type or mutant Coh2CBD-t at
various concentrations, up to a maximum of 1.3 lM), immediately
followed by the addition of dockerin solution (100 ll of XynDocS-t to
a final concentration of 47 nM). Dilutions of the competitor cohesins
were carried out in TrisNC buffer containing BSA, to maintain a
constant protein concentration. After incubation for 2.5 h, the wells
were washed five times, and the amount of dockerin bound to the
coating cohesin was detected by means of the fused-xylanase activity,
as described above.
Results were expressed as percentage of binding, derived from the
mean optical density values of five repetitions for each competitor
concentration (percentage of binding ¼100 optical density of the test
competitor concentration/optical density without competitor). Data
were analyzed using a 4-parameter fit in Grafit 5 software [20,21].
3. Results and discussion
The cohesin–dockerin interaction is the molecular adhesive
that defines and secures the cellulosome complex. To elucidate
the structural basis behind the tenacious interaction, a com-
bined bioinformatics-mutagenesis approach was exploited.
Based on amino acid sequence alignment of dockerins with
divergent specificities, we previously predicted a group of
dockerin residues that would serve as cohesin-recognition
codes. Site-directed mutagenesis was used to validate the pre-
diction, and several amino acids located on the duplicated
segments of the dockerin domain were indeed proved to be
important for the binding specificity [22,23].
In this work, a similar approach was applied to the com-
plementary module – the cohesin, to reveal the role of its in-
teracting residues and their contribution to binding and
specificity. The residues chosen for this study were based on
cohesin sequence alignment (Fig. 1) and the superposition of
related cohesin structures from two species – cohesins from C.
thermocellum and C.cellulolyticum. The amino acids impor-
tant for the binding process were assumed to be surface resi-
dues, conserved within one species but dissimilar between the
divergent species. On the basis of these criteria, various posi-
tions on the surface of the test cohesin from C. thermocellum
were subjected to site-directed mutagenesis, in which the des-
ignated residues were replaced with their counterparts from the
cohesin of C.cellulolyticum. The mutated positions were
mainly located on the conserved 8,3,6,5-face of the cohesin and
included N32, D34, V36, D65, V76, A80, and D114, as well as
replacement of the small loop connecting strands 5 and 6
(mutant 28). Several mutations (mutant 20) were also designed
to test the previously proposed involvement of residues at the
‘‘crown’’ of the cohesin [24]. The mutated genes contained
single or combined mutations, and the gene products were
overexpressed, purified, and tested for their binding specificity
and affinity.
To test the interactions of the mutated proteins, a simple
direct enzyme-linked interaction assay (ELIA) was developed.
In the non-competitive form of this assay, a cohesin solution is
used to coat microtiter plates and is allowed to interact with an
enzyme-linked dockerin solution. The enzymatic activity ob-
tained after the appropriate washings provides a direct indi-
cation of cohesin–dockerin interaction. The dockerin-fused
enzyme was xylanase T6 from Geobacillus stearothermophilus,
which is known for its exceptionally high propensity towards
expression in E. coli host cell systems [15]. Additional desirable
196 T. Handelsman et al. / FEBS Letters 572 (2004) 195–200
Document Summary
Tal handelsmana, yoav barakb, david nakarb, adva mechalyb, raphael lamedc, Received 25 may 2004; revised 15 july 2004; accepted 16 july 2004. Abstract the cohesive cellulosome complex is sustained by the high-a nity cohesin dockerin interaction. 276 (2001) 9883], we demonstrated that a single. Thr-to-leu replacement in the clostridium thermocellum dock- erin component di erentiates between non-recognition and high- a nity recognition by the interspecies rival cohesin from: cellulolyticum. In this report, we show that a single asp-to- Asn substitution on the cohesin counterpart also disrupts normal recognition of the dockerin. The asp34 carboxyl group of the cohesin appears to play a central role in the resultant hydrogen- bonding network as an acceptor of two crucial hydrogen bonds from ser45 of the dockerin domain. The results underscore the fragile nature of the intermolecular contact interactions that maintain this very high-a nity protein protein interaction. 2004 published by elsevier b. v. on behalf of the federation of.
